An Outbreak of Cadaver-Acquired Chickenpox in a ...

BRIEF REPORT

An Outbreak of Cadaver-Acquired Chickenpox in a Health Care Setting Navin Paul1 and Mini E. Jacob2 Departments of 1Medicine and Infectious Diseases and 2Community Health, Christian Medical College, Vellore, India (See the editorial commentary by Wilson on pages 602–3)

Infection due to the varicella-zoster or chickenpox virus is among the most commonly infectious human diseases, with secondary attack rates nearing 90% [1]. Clinically evident infection generally imparts lifelong immunity against exogenous reinfection. The majority of primary chickenpox virus infections occur in childhood. However, improved access to healthcare and early contact isolation of infected patients have resulted in an expanding pool of susceptible adults who lack childhood exposure to the virus [2]. Adult and immunocompromised subjects with chickenpox tend to have a more severe disease course than do children, and they are more likely to be hospitalized. They serve as potential sources of infection to nonimmune healthcare workers (HCWs) involved in their care. Hospital authorities try to minimize this risk of health care transmission by assigning HCWs with prior documented chickenpox to such patients, and by vaccinating nonimmune HCWs. Although such protocols are diligently followed in inpatient and ambulatory facilities, they may not be as rigorously adhered to in the autopsy and embalming rooms. We report an unusual outbreak of chickenpox at our center in which infection was acquired from a cadaver during the course of an autopsy. Case reports. The index case patient was a 36-year-old man who received a renal allograft and who was hospitalized in our Received 17 March 2006; accepted 11 May 2006; electronically published 27 July 2006. Reprints or correspondence: Dr. Navin Paul, Dept. of Medicine, Bridgeport Hospital, 267 Grant St., Bridgeport, CT 06610 ([email protected]). Clinical Infectious Diseases 2006; 43:599–601 2006 by the Infectious Diseases Society of America. All rights reserved. 1058-4838/2006/4305-0009$15.00

BRIEF REPORT • CID 2006:43 (1 September) • 599

Downloaded from http://cid.oxfordjournals.org/ at Boston University on October 9, 2016

The varicella-zoster virus (the chickenpox virus) is highly infectious and may affect nonimmune health care professionals. We report an outbreak of chickenpox in our hospital in which the source of infection was a cadaver. The spread of infection occurred in the course of an autopsy.

tertiary care center with extensive mucocutaneous vesicular eruptions for 7 days and fever for 3 days. He had been receiving daily immunosuppressive therapy with azathioprine (150 mg) and prednisone (10 mg) after undergoing renal transplantation 10 years earlier. No past history of chickenpox was elicited. Examination revealed tachypnea, blood pressure of 110/70 mm Hg, and temperature of 36.9C. Small, tense vesicles, some of which were hemorrhagic, were distributed all over the body. Tzanck smear was positive for multinucleate giant cells. Investigations suggested hemoconcentration, disseminated intravascular coagulation, and an elevated liver enzyme level (11500 U/L), implying disseminated varicella-zoster virus infection with organ involvement. The baseline serum creatinine of 1.6 mg/dL had worsened to 3.9 mg/dL at hospitalization. ELISA was performed to detect antibodies to HIV, and the results were negative. Despite prompt initiation of parenteral acyclovir and broad-spectrum antibiotic treatment, he developed worsening hypotension refractory to maximal doses of dopamine. Twelve h after hospitalization, the patient had a cardiorespiratory arrest that was unresponsive to resuscitatory measures, and he was pronounced dead. The body was moved to the mortuary for refrigeration while awaiting autopsy. Autopsy was performed 12 h after death. The procedure was attended by 22 HCWs: a pathologist who performed the procedure, 2 assistants, and 19 third-year medical students undergoing a pathology rotation. Masks and plastic aprons were worn by all attendees. Although the pathologist and the assistants had reported past history of chickenpox, the medical students were not questioned regarding prior chickenpox or vaccination. Taking into account the danger of splash of infective body fluids during autopsy, the students were warned to remain some distance from the autopsy table; however, this recommendation went largely unheeded. A brief interruption occurred while demonstrating the sectioning of the liver, when blood splashed onto some attendees. The procedure was resumed without any attempt to identify those who were exposed or to wash the blood off of those who received the splash. The first case of chickenpox among those who attended the autopsy was reported 12 days later. Three more cases emerged over the next 6 days. All 4 attendees who developed chickenpox reported experiencing a splash of blood from the cadaver onto an area of their exposed, unbroken skin during the autopsy. None reported mucosal exposures. None of the unaffected attendees reported being splashed. Although 2 of the affected subjects

600 • CID 2006:43 (1 September) • BRIEF REPORT

infective agents after death of the host might contribute to such instances. In the case of infections in which data regarding postmortem viability of causative organisms is unavailable, a good approach might be to follow the precautions that would be taken with a live patient with the disease. Apart from its lessons regarding infection control in the autopsy room, our series also seems to supply some clues about the transmission and infectivity of the varicella-zoster virus. Chickenpox infection is believed to spread through respiratory aerosols or direct contact with skin sores [5]. Despite its high secondary attack rates among family members of diseased persons, infectivity is modest among HCWs with occupational exposure [6]. In our series, contact with body fluids was more efficient in transmitting chickenpox than respiratory aerosol, as is evident from the absence of clinical infection among the secondary contacts with only a respiratory source of infection. Thus, the high secondary attack rates among family members may result from amplified opportunities for direct contact with infective vesicular fluid. The clinical features of patients in our series were similar to textbook descriptions of chickenpox, suggesting that the route of infection may not play a role in symptomatology. Reinfection with the varicella-zoster virus is believed to result from loss of specific immunity, usually due to immunosuppression or advanced age [7, 8]. Our 2 subjects who experienced varicella-zoster virus reinfection were both aged !30 years and were immunocompetent. We hypothesize that a relatively high percutaneous dose of exposure to the infectious agent might have been responsible. A large enough infectious dose of the virus may thus overwhelm varicella-zoster virus–specific immunity and lead to clinically apparent reinfection. However, this observation should be tempered by the fact that history of prior chickenpox in our subjects was determined solely on the basis of their recollection of the disease, and not on serological criteria. Our report is limited by the absence of serological or virological data. However, in real-world practice, chickenpox is a clinical diagnosis, and microbiological confirmation is sought only in cases with atypical features. The common source of exposure in our subjects and the development of skin lesions typical of chickenpox within the expected incubation period make an alternative diagnosis highly unlikely. In conclusion, direct contact with body fluids seems to be more efficient in transmitting the chickenpox virus than aerosols of infective material. Reinfection could result from a sufficiently high dose of exposure. Clinical features do not seem to be influenced by the route of acquisition of infection. In view of recent reports of incomplete protection with chickenpox vaccination, adequate preventive measures and history of past exposure to the varicella-zoster virus should be used in conjunction with vaccine to interrupt transmission in health


reported having had chickenpox in the past, the other 2 denied such history. The mean age of the infected HCWs was 22.5 years (range, 20–28 years). The mean incubation period was 15 days (range, 12–18 days). All 4 subjects spent their childhoods in the state of Kerala. Three were hospitalized in the hospital’s isolation facility to prevent further spread of infection among the student population, whereas the fourth preferred to be confined to home. The clinical features of the 3 hospitalized subjects were as follows: mean time to last crust, 12.3 days; appearance of vesicles before fever, 2 of the 3 patients; appearance of vesicles on trunk before face, 2 of the 3 patients; body ache, all 3 patients; pruritus, all 3 patients; mucosal involvement, 2 of the 3 patients; and conjunctival congestion, 1 of the 3 patients. The last spike in temperature followed the last crop of vesicles by 24 h in all subjects. None had any coexisting medical illnesses. Acyclovir therapy was not instituted in any subject. All recovered without major complications. The 3 hospitalized subjects reported close contact (defined as having spent at least 2 h together in an enclosed space during the 2 days before onset of vesicles) with 8 identifiable persons. Five of these 8 subjects reported no prior history of chickenpox. None had received a chickenpox vaccination. Although none acquired chickenpox during the outbreak, 4 of these 5 secondary contacts developed the disease at some time over the next 5 years, implying that they lacked immunity to the varicella-zoster virus at the time of their exposure to the primary contacts. Discussion. Our unique case series demonstrates the importance of assessing HCWs’ susceptibilities to the varicellazoster virus before exposing them to situations with increased risk of infection acquisition. Multiple instances of disregarded protocols led to disease among our subjects. Simple measures, such as obtaining history of prior chickenpox infection or prompt identification and care of attendees exposed to a splash of infected fluids, might have helped forestall this outbreak. Acquisition of chickenpox is an occupational hazard among HCWs with no immunity to the infection. Although administration of a chickenpox vaccine was considered to be a promising solution, recent reports have suggested that its real-life efficacy may not be equivalent to that which was reported in initial studies [3]. This further emphasizes the important role for good prevention measures in averting outbreaks of chickenpox among HCWs. The embalming and autopsy areas of the hospital are at risk of being overlooked when infection-control measures are implemented. Although we could not find any previous reports of varicella-zoster virus transmission in the autopsy room, there have been prior accounts of other infectious disease transmission in similar circumstances due to inadequate preventive measures [4]. Deficient knowledge regarding the viability of

care setting. Such protocols for prevention should be as rigorously followed in the autopsy and embalming rooms, as in other parts of the hospital. Acknowledgments Potential conflicts of interest. All authors: no conflicts.

References 1. Ross AH. Modification of chicken pox in family contacts by administration of gamma globulin. N Engl J Med 1962; 267:369–76. 2. Whitley RJ. Varicella-zoster virus. In: Mandell GL, Bennett JE, Dolin R, eds. Principles and practice of infectious diseases, 5th ed. New York: Churchill Livingstone, 2000:1580–6.

3. Galil K, Lee B, Strine T, et al. Outbreak of varicella at a day care center despite vaccination. N Engl J Med 2002; 347:1909–15. 4. Sterling TR, Pope DS, Bishai WR, Harrington S, Gershon RR, Chaisson RE. Transmission of Mycobacterium tuberculosis from a cadaver to an embalmer. N Engl J Med 2000; 342:246–8. 5. Parmet S, Lynm C, Glass RM. JAMA patient page: chickenpox. JAMA 2005; 294:866. 6. Langley JM, Hanakowski M. Variation in risk for nosocomial chickenpox after inadvertent exposure. J Hosp Infect 2000; 44:224–6. 7. Schoub BD, Blackburn NK, Johnson S, McAnerney JM, Miller B. Low antibody avidity in elderly chickenpox patients. J Med Virol 1992; 37: 113–5. 8. Junker K, Avnstorp C, Nielsen CM, Hansen NE. Reinfection with varicella-zoster virus in immunocompromised patients. Curr Probl Dermatol 1989; 18:152–7.


BRIEF REPORT • CID 2006:43 (1 September) • 601